PROJECT SUMMARY More than 30 autoimmune type 1 diabetes (T1D) susceptibility loci (termed Idd) have been identified in the nonobese diabetic (NOD) mouse, a spontaneous animal model for the human disease. Among those, the Idd9.3 locus has been mapped to a 1.22 Mb region on Chromosome 4 containing 17 protein-coding and 12 non-coding genes, of which Tnfrsf9 (encoding CD137) is the top candidate. The C57BL/10 (B10)-derived Idd9.3 confers T1D resistance, whereas the NOD-derived interval contributes to disease development. Due to the tight linkage of the region, congenic strains that can be used to further dissect the Idd9.3 locus have not been made available to more precisely determine if Tnfrsf9 is the underlying gene. Human genome wide association studies have linked TNFRSF9 to several autoimmune diseases, and CD137 functionally interacts in immune pathways with known human T1D genes (e.g., TNFAIP3). Thus, it is important to further study the role of CD137 in T1D. It has been previously reported that NOD-derived CD137 is hypofunctional compared to B10 CD137. On the other hand, we demonstrate that CD137-deficient NOD mice, generated by zinc-finger nuclease (ZFN) mediated mutagenesis, are resistant to T1D. We further show here that CD137 expression in CD4 and CD8 T-cells respectively suppresses and promotes T1D development in NOD mice, but CD137 deficiency dominantly affects CD8 T-cells leading to disease protection. Thus, T1D development in NOD mice carrying different alleles of Tnfrsf9 (B10, NOD, or the null allele) is influenced by the combined effects of its distinct functions in different cell types. A better approach is needed to conclusively determine if Tnfrsf9 is the Idd9.3 gene and to compare the function of NOD and B10 alleles in CD4 versus CD8 T-cells. The ability of the ZFN technology to specifically knock out a gene in mouse strains lacking germ-line transmittable embryonic stem cells has also allowed us to target Tnfrsf9 in the NOD mice congenic for the B10 Idd9.3 region. Our goal in this application is to definitively determine if Tnfrsf9 is the Idd9.3 underlying gene by establishing a pair of genetically identical strains where only one parental allele of Tnfrsf9 (NOD or B10) is expressed, and to further examine whether allelic variation of CD137 changes its dominant mode of action in T1D.